We have recently identified and characterized two scaffold attachment factors termed SAFB1 and SAFB2, which show 75 percent homology and map together (in head-to-head orientation) on chromosome 19p13. We have discovered that the SAFBs function as growth inhibitors in breast cancer cells, and that their locus on chromosome 19p13 displays one of the highest losses of heterozygosity (78 percent) reported in clinical breast cancer. Protein expression of SAFB1/2 is lost in approximately 20 percent of breast cancers, and low expression is associated with a shorter overall survival. We have found SAFB1 mutations in breast tumors that are not present in adjacent normal tissue. Therefore, we hypothesize that SAFB1 and SAFB2 are tumor suppressors whose frequent inactivation in breast tumors results in tumorigenic and metastatic phenotypes, which we will assess using in vivo and in vitro models. While both proteins function as growth inhibitors and ER co-repressors, they also have distinct properties - both are found in the cell nucleus, but only SAFB1 is recruited to nuclear bodies under stress conditions, and only SAFB2 also appears in the cytoplasm and binds vinexin-beta. We will therefore assess which SAFB functions are essential for tumor suppression, and which functions are additive, synergistic, compensatory, or antagonistic between SAFB1 and SAFB2. Specifically, we will ask: 1) How are SAFB1/2 inactivated in human breast tumors? We will investigate loss during breast cancer evolution, and determine whether inactivation occurs through mutations and/or other epigenetic events such as hypermethylation of the intergenic bidirectional SAFB1/2 promoter. 2) What are the biological effects of neutralizing SAFB1 in vitro and in vivo? We will compare and contrast the effects of lowering (by RNAi) SAFB1 and SAFB2, of SAFB mutants identified in clinical breast cancer specimens, and of gene knockouts in mouse models (SAFB1-/- and SAFB2-/-). 3) How do SAFB1 and SAFB2 levels correlate with clinical breast cancer outcome? We will measure SAFB1 and SAFB2 levels in an arrayed breast tumor bank, and correlate the levels with survival and other known prognostic factors. Successful completion of this study, which incorporates molecular, cellular, and in vivo genetic approaches, will provide important insight into regulatory mechanisms which are critically disturbed in human breast cancer.